Media guidance applications often provide opportunities for users to rate a media asset, either as they consume the media asset or after they have consumed the media asset. Such media guidance applications also may allow the user to rate individual scenes of the media asset separately. For example, a user who enjoys comedies may rate a comedic scene in a movie very highly, but may give a low rating to a later romantic scene in the same movie. In this manner, the media guidance application may track not only the user's overall impression of the media asset, but his or her interests and preferences as they change during the consumption of the media asset.
This information about the user may be used to drive media asset recommendations. For instance, a media guidance application may track which scenes a user rates highly and recommend related media assets with similar scenes. The accuracy of such media asset recommendations may be improved by not only deriving the user's preferences from the scene-level ratings, but also by tracking the pattern of user ratings as they change over the course of the duration of the media asset.